This may be applicable to Goodwes, maybe more general. I don't know. I share it here as a lesson learned and that somebody may not have to learn in future. And maybe because somebody who really understands the Goodwe ES inverters may offer corrections & further enlightenment.

Saturday morning we had an early morning outage (not load shedding). Round about 9AM. It was a clear, sunny day, but look...

If you're not familiar with SEMS portal, the blue line is PV being drawn from the panels, green line is SOC, orange line is load. 

PV drops just as the outage begins. You can't see the blue line because it is on the zero axis, behind other lines.

Now I have for some time run my Goodwe in "Eco" mode. This allows you to set time slots during which the battery will discharge or charge. I have it set for one charge slot - 15:00 to 16:00. This so that on overcast days my battery gets an afternoon charge and I am good for the evening.

So the first thing I checked was the front panel of the Goodwe. This has an LED to show if either of the PV strings is active. It was burning a solid blue - so it could "see" both strings. I checked the Goodwe app - it showed 250 odd volts on each string, but no current.

OK... this should have led me away from the path that I was about to follow but I

1) Turned both the PV breakers off (LED went out). 
2) Checked the fuses with a multimeter - they are good.
3) Replaced the fuses, checking they were located correctly.
4) Turned the breakers both on.

Well maybe it's going to kick start something. But it didn't. The app still reported 250 odd volts and no current. The LED still burned., No PV was being drawn.

At this point I start googling for after hours PV troubleshooters. There don't see to be any. So now what?

Had I played with any settings lately? No. This is against my religion. I don't touch that stuff. Also the Freedom Won BMS will override any changes I make on the battery side (spoil sports!).

So I changed the mode from "eco" to "off grid" because, really, I am off grid.

Bingo! It starts drawing PV and charging the batteries again.

In for a penny, in for a pound. I now changed it back to "general" mode. In this mode it uses PV, battery, grid in that order. Now when the grid comes back the inverter will know and can use it if it needs to.

But now I started thinking back to another recent incident. Thursday just gone.

So now the PV is up and down and the battery is actually discharging more than it is charging. I checked the Goodwe app, and the incoming voltage is low. Alternating between mid 180s and low 200s. The Goodwe, with safety code set to "South Africa", disconnects from the grid at around 195V. It reconnects at about 200. The change I made here was to switch it to the "50Hz default" safety code. Set like this, the Goodwe doesn't care what's coming in as long as it's above 170 and it's at 50Hz. You can see the problem going away.

So I should have put two and two together here: When the inverter is in eco mode, when the time is not within a charge slot, and when it doesn't accept or get any grid input, it doesn't take any PV, no matter how much is available.

Later in the day, with voltage back but still low, I changed safety code to "South Africa" and inverter mode to "eco". Straight away the PV drops to zero (inverter not getting any grid power).

I then put the inverter back into "general" mode and turned off the main breaker for the property. The drawing of PV was unaffected.

This was still strange to me, because I had the inverter stuck in eco mode with that single 15:00 to 16:00 charge slot for over a year. But looking back through the history on SEMS I found smoking guns. Like this one

 

Conclusion: If the following conditions are in place
1) Inverter in eco mode
2) The time of day is outside of a charge time as per eco mode
3) No grid coming into inverter (for whatever reason)

Then the inverter will not draw any PV. Possibly until a charge cycle/time slot is entered.

So I am leaving my Goodwe inverter in general mode, unless I have to charge from grid.

I had left it stuck in eco mode because you can change modes remotely, it's a security blanket, and because I was too lazy to change the mode every day.

 

 

 

Edited July 2, 20232 yr by Bobster.
correct the screen shots

ECO mode is designed for people with TOU tariffs to arbitrage (buy cheap and sell when the price is high). If you are in SA and use ECO mode, be prepared for issues like this.

OK... I don't think the mode is the issue. This puzzled me. I take @P1000's point, but I'd had the Goodwe in ECO mode for over 2 years and everything worked OK.

Now today I end up again with inverter not drawing from PV nor grid, and this is with it set in General mode, and grid at 203 V - not ideal, but the Goodwe should still accept this.

So it happened again this morning. Forewarned is forearmed and I was looking out for it.

About 7:00 I started getting solar coming in, but not a lot obviously.

Inverter was in general mode with safety code set to South Africa. And it's been operating happily in that configuration all week.

8:00 load shedding started.

8:00 the inverter stops drawing from the panels, even though it is connected to both strings, and can see 250V plus on each string.

I change the mode from general to off-grid and this kick starts something and the system starts drawing PV again.

I change the mode from off-grid back to general and it carries on happily using PV.

I tell the household they can use the kettle.

The installer is no help. After all this time I don't expect free service, but it would be nice to get a response. Their Goodwe expert has moved to some other venture and is now not available to me 😞 

I am going to log a call with Goodwe support desk and see what happens.

I'm also going to start looking around for somebody who will trouble shoot system they didn't install (this is out of the installer's warranty, so there is no problem there) and who, hopefully, knows the Goodwe ES inverters.

Not directly responding to the topic of this thread.
 

But for years I have been running in either General mode (Stage 3 or less) and Eco Mode (stage 4 up) with no issues. 
 

Then Goodwe began to “enhance” their SEMS platform, stop PV Master app and switchbto SolarGo app. 
 

And for the last 6 months nothing works as expected on a constant basis

 

Conclusion: If the following conditions are in place
1) Inverter in eco mode
2) The time of day is outside of a charge time as per eco mode
3) No grid coming into inverter (for whatever reason)

Then the inverter will not draw any PV. Possibly until a charge cycle/time slot is entered.

So I am leaving my Goodwe inverter in general mode, unless I have to charge from grid.

I had left it stuck in eco mode because you can change modes remotely, it's a security blanket, and because I was too lazy to change the mode every day.

 

 

 

Hi @Bobster.

I use Economy Mode and what you say is very strange. Although I hope I haven't misunderstood you.
Since I export power to the grid, to prevent my batteries from eventually discharging to the grid, I never have a discharge period with economy mode during high solar production hours of the day.
But since I discovered what I believe to be a bug in the ES 5048 firmware when a VAC failure condition exists, I tested its conclusions as it is important information that I could incorporate into my software development.
I set the economy mode with a DISCHARGE period in the presence of a large amount of PV (around noon) and proceeded to cut the Grid.
There was no change. The inverter automatically supplies from PV all the power needed for essential loads.

 

 

Hi @Bobster.

I use Economy Mode and what you say is very strange. Although I hope I haven't misunderstood you.
Since I export power to the grid, to prevent my batteries from eventually discharging to the grid, I never have a discharge period with economy mode during high solar production hours of the day.
But since I discovered what I believe to be a bug in the ES 5048 firmware when a VAC failure condition exists, I tested its conclusions as it is important information that I could incorporate into my software development.
I set the economy mode with a DISCHARGE period in the presence of a large amount of PV (around noon) and proceeded to cut the Grid.
There was no change. The inverter automatically supplies from PV all the power needed for essential loads.

 

Hi @Cef.

The problem seems to have gone away.

There was a problem in my area (my street and surrounding streets) with low grid voltage. When I say low, I mean down in the 170V range when we are supposed to have 230V. For a couple of months we didn't see 230, would sometimes see 210. We raised this with the municipality, and were able to give them exact locations, times, and details of how the voltage was measured. 

About a month ago now, the power suddenly went out at about 4pm. This was unscheduled. It was only down for about 5 minutes, and clearly the municipality changed something because when it came back on we were getting a nice 235V. Since then the voltages have been in a much more acceptable range, and my inverter has started behaving itself again.

And since that incident, my system has worked predictably and as it should. And as we know, this is not a situation in which you can learn much.

So I now have to wait for another bout of strange behaviour and see if this appears to be linked to very low voltage.
 

Edited September 13, 20232 yr by Bobster.

 

Hi @Cef.

The problem seems to have gone away.

There was a problem in my area (my street and surrounding streets) with low grid voltage. When I say low, I mean down in the 170V range when we are supposed to have 230V. For a couple of months we didn't see 230, would sometimes see 210. We raised this with the municipality, and were able to give them exact locations, times, and details of how the voltage was measured. 

About a month ago now, the power suddenly went out at about 4pm. This was unscheduled. It was only down for about 5 minutes, and clearly the municipality changed something because when it came back on we were getting a nice 235V. Since then the voltages have been in a much more acceptable range, and my inverter has started behaving itself again.

And since that incident, my system has worked predictably and as it should. And as we know, this is not a situation in which you can learn much.

So I now have to wait for another bout of strange behaviour and see if this appears to be linked to very low voltage.
 

Hahaha! I tempted fate, and this morning fate came knocking.

We had load shedding, a 4 hour shed starting at 4:00. Lovely sunny morning and PV production was good early on.

Then the power came back on. And the voltage on start up was low.

Straight away, without any eclipse of the sun, hadedas congregating on the panels, clouds, etc, the PV production dropped to just enough to track the load. Grid voltage on restoration was about 185V.

Now there is still an unknown quantity in this. Me! I just happened to notice this and went into my usual pattern of changing the operation mode and etc. It may be that if I did nothing, normal service would have resumed anyway. But how long should I wait for that? 

And this didn't happen yesterday when we we also had an early morning load shed and the power came back just after 8. But of course because there was no problem (or no problem observed) I didn't check anything and so can't say what the grid voltage was.

@Bobster.

I have same inverter .... can't speak to your scenario specifically (and I haven't read through this entire thread), but I have noticed that when there's big power shifts (switching big geyser off for example) then PV production drops drastically (to track load, or even less meaning slight import from Eskom), and then takes a while (maybe 5min) to ramp up production again (and/or pull from batteries rather than grid). 
I vaguely remember once having Eskom come back and it doing same as you describe (running general mode), but that could be a false memory

Point is, next time it happens, let it stand for 10min and see if it "catches up" again. I do find it kinda annoying the way it seems to initially "shift to grid" on big power shifts (big load off or on, or as you state load shedding ends), maybe its to do with not hammering the battery this way and then that. In the big scheme of things however, for my own use, its a small annoyance rather than an actual issue...

 

@Bobster.

Point is, next time it happens, let it stand for 10min and see if it "catches up" again. I do find it kinda annoying the way it seems to initially "shift to grid" on big power shifts (big load off or on, or as you state load shedding ends), maybe its to do with not hammering the battery this way and then that. In the big scheme of things however, for my own use, its a small annoyance rather than an actual issue...

Thanks, and you're right. This is exactly what I need to do. I've noticed the shifting that you mention, though that's usually short lived. It seems to take a small but detectable period of time to increase PV when it needs it. 

So now I wait it for it to hit another wobbly. Which will probably be when I'm not at home and so can't do anything about it. Then I resist my urge to intervene.

 

Straight away, without any eclipse of the sun, hadedas congregating on the panels, clouds, etc, the PV production dropped to just enough to track the load. Grid voltage on restoration was about 185V.

 

In the VAC Failure condition (Pv Master registers this error with the code: 131072) there is nothing that surprises me. In my location the energy is very stable and is within the voltage and frequency ranges of the country code. Unfortunately, I cannot reproduce both a FAC and a VAC condition to draw definitive conclusions.
However, from a message from @paul99I was able to observe how in VAC condition, the information that the Goodwe provides in one of its sensors (the one for Non-ESSENTIAL loads) is wrong. Without a doubt that is a bug in the Goodwe firmware. The information shown by PV master is not divided into essential and non-essential loads.
It is possible that in this condition of VAC Failure, and very possibly FAC Failure (Code: 536870912) the PV energy is =0. Hopefully Paul can tell us more about this.

And it happened again! 

Monday there was a big storm here. Monday afternoon myself and my neighbours on each side started noticing low voltages. Like low. Like 160s, sometimes 150s. During the night it would creep up, hitting about 240 at 3am (don't ask). Meanwhile the inverter is reporting "VAC failure" errors, and refuses to connect to the grid even when the reported voltage is in range. Interestingly the trouble shooting guides I could find said to check the neutral connection when this happens. I didn't. Because I'm tired of getting that good old buzzing feeling when I start fiddling with anything more complicated than a 16A plug.

And this morning, with the grid down in the 160s and 170s, the inverter resumes it's pattern of drawing just enough solar to service the load - on a nice bright morning. I fixed this once by changing settings via SolarGo. It worked. For a while. Then another VAC failure error and we're back at square one. This time I resisted the urge and went and did some planking and had breakfast. Slowly the PV picked up.

End of the tale: City Power came knocking, said they were investigating reports of low voltage in 3 houses, mine being one. They first asked me to turn off any inverters in the house then did a stress test. Measure the grid voltage at the meter, then I turn a whole load of stuff on and they see what happens. They then moved this test to.... 

Long story short: They eventually found a fuse on the neutral line at the substation which had not been located into the holder properly. They pushed this in, and suddenly everybody starts getting a nice, steady 230V. And my inverter started behaving itself again.

The worrying scenario for me had been that there was grid voltage, PV was low because of the stormy weather, and because the inverter refused to connect despite the voltage being in range (sometimes) I couldn't get charge into my batteries. I managed to limp through with PV getting some charge into the batteries, and me not having hot showers. Since the dodgy neutral in the substation was fixed, the inverter is as happy as a pig in the proverbial and normal service has resumed.

3 hours ago, Bobster. said:

And it happened again! 

Monday there was a big storm here. Monday afternoon myself and my neighbours on each side started noticing low voltages. Like low. Like 160s, sometimes 150s. During the night it would creep up, hitting about 240 at 3am (don't ask). Meanwhile the inverter is reporting "VAC failure" errors, and refuses to connect to the grid even when the reported voltage is in range. Interestingly the trouble shooting guides I could find said to check the neutral connection when this happens. I didn't. Because I'm tired of getting that good old buzzing feeling when I start fiddling with anything more complicated than a 16A plug.

And this morning, with the grid down in the 160s and 170s, the inverter resumes it's pattern of drawing just enough solar to service the load - on a nice bright morning. I fixed this once by changing settings via SolarGo. It worked. For a while. Then another VAC failure error and we're back at square one. This time I resisted the urge and went and did some planking and had breakfast. Slowly the PV picked up.

End of the tale: City Power came knocking, said they were investigating reports of low voltage in 3 houses, mine being one. They first asked me to turn off any inverters in the house then did a stress test. Measure the grid voltage at the meter, then I turn a whole load of stuff on and they see what happens. They then moved this test to.... 

Long story short: They eventually found a fuse on the neutral line at the substation which had not been located into the holder properly. They pushed this in, and suddenly everybody starts getting a nice, steady 230V. And my inverter started behaving itself again.

The worrying scenario for me had been that there was grid voltage, PV was low because of the stormy weather, and because the inverter refused to connect despite the voltage being in range (sometimes) I couldn't get charge into my batteries. I managed to limp through with PV getting some charge into the batteries, and me not having hot showers. Since the dodgy neutral in the substation was fixed, the inverter is as happy as a pig in the proverbial and normal service has resumed.

Sorry that you have to go through all of this. 

I must say I LIKE the fuse in the neutral as indicated. 

On 2024/01/31 at 7:31 PM, Scorp007 said:

Sorry that you have to go through all of this. 

I must say I LIKE the fuse in the neutral as indicated. 

It interested me that the literature I could find was telling me to check the neutral, and that it did turn out to be a problem with the neutral, just not the inverter's connection to neutral. Since City Power correctly located the fuse, the Goodwe has behaved itself as expected.

2 hours ago, Bobster. said:

It interested me that the literature I could find was telling me to check the neutral, and that it did turn out to be a problem with the neutral, just not the inverter's connection to neutral. Since City Power correctly located the fuse, the Goodwe has behaved itself as expected.

The thing is a neutral may not have a fuse as you NEVER break a neutral unless the live is also interrupted the same time. Thus only MCBs can be used as a double pole to break neutral and live. 

Due to the neutral carrying the fault current back to earth the above has merit being part of SANS. 

1 hour ago, Scorp007 said:

The thing is a neutral may not have a fuse as you NEVER break a neutral unless the live is also interrupted the same time. Thus only MCBs can be used as a double pole to break neutral and live. 

Due to the neutral carrying the fault current back to earth the above has merit being part of SANS. 

 

 

On 2024/01/31 at 3:39 PM, Bobster. said:

They eventually found a fuse on the neutral line at the substation which had not been located into the holder properly

What they should have said that someone working on the transformer forgot to replace the neutral link before the transformer was energized.

1 hour ago, TaliaB said:

 

 

What they should have said that someone working on the transformer forgot to replace the neutral link before the transformer was energized.

Normally there is only a link on the transformer. In this case of @Bobster. with 3 houses affected it could be at a kiosk?

Our kiosks have a 125A incomer MCB that splits to the different houses that are fed with a 80A. Then 60A at the side of the house feeding the main DB. 

1 hour ago, TaliaB said:

 

 

What they should have said that someone working on the transformer forgot to replace the neutral link before the transformer was energized.

OK... my understanding, because we recently had this in our street, is that an open circuit on the neutral leads to people getting shocks off of taps and nasty incoming voltages. The guy said "fuse" to me, and he said it was on the neutral (I hadn't mentioned anything about a "neutral" to him). So is it possible that the wild fluctuation of voltage (usually fluctuating low, but sometimes high) and my inverter's unhappiness was caused by a bad connection (or high resistance) on the neutral line because this link wasn't properly in place?
 

43 minutes ago, Scorp007 said:

Normally there is only a link on the transformer. In this case of @Bobster. with 3 houses affected it could be at a kiosk?

If I understood the City's electrian then no. He first said, well, it could be something in your house, we have to check. So he asked me to turn everything off to see if the voltage at the meter went back up. Then he asked me to turn some big loads on to see if that pulled the voltage down. He then disconnected the other two properties at the top of the pole so that only I was connected and repeated those tests.

Then he disconnected me, reconnected one of my neighbours and did the same stress tests there.

Then he disconnected all three properties from the pole and measured there a few minutes and saw fluctuation and a generally low voltage. AIUI he was able to test to satisfy himself that this was incoming.

Then he went looking for branches upstream of the pole fouling the cable. Then he went off to check the substation. By the time he got back from there I was sitting happily at 299  229 V, but he repeated the stress test to be sure. Said he didn't want to have to come back tomorrow.

Edited February 2, 20242 yr by Bobster.